PROJECT SUMMARY Lyme disease is the most frequently reported vector-borne disease in the U.S., with 300,000 cases estimated to occur annually. Current diagnosis of Lyme disease is based on recognition of an erythema migrans (EM) skin lesion or positive two-tiered serological (antibody detection) testing in a patient with consistent clinical signs and tick exposure in areas where Lyme disease occurs. It is estimated that 20 to 30% of patients do not present with an EM, and the majority of patients do not recall a tick bite. Moreover, serologic tests are dependent on the host humoral immune response and lack sensitivity in early Lyme disease (only 29-40% of patients with EM are seropositive). Serological reactivity also may persist for years following antibiotic treatment and resolution of symptoms. These limitations and the need for early diagnosis to facilitate rapid therapeutic intervention provide strong rationale for the development of new Lyme disease diagnostic tests. We have undertaken a novel approach of applying serum metabolomics to develop small molecule biosignatures that can be exploited as a diagnostic test for early Lyme disease. These efforts resulted in a published candidate biosignature that provided a sensitivity of 88% for early Lyme disease with a specificity of 95% for healthy controls and 93% for other disease control populations. This approach was also able to differentiate early Lyme disease from Southern Tick Associated Rash Illness (STARI), an illness with an EM-like skin lesion and similar non-specific symptoms of early Lyme disease, and correctly classified these two patient groups with 98% accuracy for Lyme disease and 89% accuracy for STARI. Our preliminary data now provides strong evidence that metabolic profiles can differentiate early LD from other tick transmitted diseases and distinguish between the various manifestations of early Lyme disease (i.e., early localized versus early disseminated disease). Under this proposal, the expertise of biochemists, microbiologists, mathematicians, statisticians and infectious disease clinicians will be combined to perform studies that will significantly advance our previous efforts. Specifically, we hypothesize that it is possible to create a diagnostic metabolic profile that accurately distinguishes early Lyme disease patients from non-Lyme disease patients, and that can be applied in a clinical laboratory, Importantly the non-Lyme disease patient group are those individuals suspected of Lyme disease (patients who present for medical care and who undergo diagnostic testing for Lyme disease, but are not diagnosed with Lyme disease), as well as those with other tick transmitted diseases. Our proposed efforts take into account the heterogeneous symptoms of early Lyme disease and the non-Lyme disease populations. The goal of diagnostic development will be facilitated through the application of state-of-the-art mathematical modeling and well-characterized prospectively and retrospectively collected sera. Metabolites with the greatest discriminatory value for early Lyme disease will be structurally elucidated and characterized to facilitate implementation of a multianalyte multiple reaction monitoring assay as a platform for early Lyme disease diagnosis. Metabolic pathways associated with early Lyme disease will be elucidated to establish a biological rationale that support metabolic profiling as a diagnostic method for early Lyme disease.